I've started to see my Exit node restart itself. There is a warning in the log pointing to a problem with OpenSSL but I don't know how to proceed.
Jun 21 03:07:03.000 [notice] Tor 0.2.9.10 (git-1f6c8eda0073f464) opening new log file. Jun 21 03:07:03.619 [warn] OpenSSL version from headers does not match the version we're running with. If you get weird crashes, that might be why. (Compiled with 1000105f: OpenSSL 1.0.1e 11 Feb 2013; running with 1000105f: OpenSSL 1.0.1e-fips 11 Feb 2013). Jun 21 03:07:03.647 [notice] Tor 0.2.9.10 (git-1f6c8eda0073f464) running on Linux with Libevent 2.0.21-stable, OpenSSL 1.0.1e- ips and Zlib 1.2.7. Jun 21 03:07:03.647 [notice] Tor can't help you if you use it wrong! Learn how to be safe at https://www.torproject.org/download/download#warning Jun 21 03:07:03.647 [notice] Read configuration file "/usr/share/tor/defaults-torrc". Jun 21 03:07:03.647 [notice] Read configuration file "/etc/tor/torrc". Jun 21 03:07:03.650 [notice] Based on detected system memory, MaxMemInQueues is set to 1464 MB. You can override this by setting MaxMemInQueues by hand. Jun 21 03:07:03.650 [warn] Tor is running as an exit relay. If you did not want this behavior, please set the ExitRelay option to 0. If you do want to run an exit Relay, please set the ExitRelay option to 1 to disable this warning, and for forward compatibility. Jun 21 03:07:03.650 [notice] Opening Socks listener on 127.0.0.1:9050 Jun 21 03:07:03.650 [notice] Opening Control listener on 127.0.0.1:9051 Jun 21 03:07:03.650 [notice] Opening OR listener on 0.0.0.0:8081 Jun 21 03:07:03.651 [notice] Opening Directory listener on 0.0.0.0:8080 Jun 21 03:07:03.000 [notice] Not disabling debugger attaching for unprivileged users. Jun 21 03:07:03.000 [notice] Parsing GEOIP IPv4 file /usr/share/tor/geoip. Jun 21 03:07:03.000 [notice] Parsing GEOIP IPv6 file /usr/share/tor/geoip6. Jun 21 03:07:03.000 [notice] Configured to measure statistics. Look for the *-stats files that will first be written to the data directory in 24 hours from now. Jun 21 03:07:03.000 [warn] Fixing permissions on directory /var/lib/tor Jun 21 03:07:03.000 [notice] We were built to run on a 64-bit CPU, with OpenSSL 1.0.1 or later, but with a version of OpenSSL that apparently lacks accelerated support for the NIST P-224 and P-256 groups. Building openssl with such support (using the enable-ec_nistp_64_gcc_128 option when configuring it) would make ECD H much faster.
thanks for running an exit.
Alan:
I've started to see my Exit node restart itself.
are you talking about tor daemon / process restarts or entire host restarts?
I guess you run a RPM based distro. I do not think that the openssl message has anything to do with the restarts. What does your syslog (/var/log/messages) say? any out-of-memory kills?
thanks for running an exit.
Alan:
I've started to see my Exit node restart itself.
are you talking about tor daemon / process restarts or entire host restarts?
I guess you run a RPM based distro. I do not think that the openssl message has anything to do with the restarts. What does your syslog (/var/log/messages) say? any out-of-memory kills?
syslog says it failed but didn't give a reason.
Jun 21 03:07:01 vps110165 systemd: Reloaded Anonymizing overlay network for TCP (multi-instance master). Jun 21 03:07:01 vps110165 systemd: Reloaded Anonymizing overlay network for TCP. Jun 21 03:07:02 vps110165 systemd: tor.service: main process exited, code=exited, status=1/FAILURE Jun 21 03:07:02 vps110165 systemd: Unit tor.service entered failed state. Jun 21 03:07:02 vps110165 systemd: tor.service failed. Jun 21 03:07:03 vps110165 systemd: tor.service holdoff time over, scheduling restart. Jun 21 03:07:03 vps110165 systemd: Starting Anonymizing overlay network for TCP...
On 22 Jun 2017, at 04:38, Alan tor-relay@clutterbuck.uk wrote:
thanks for running an exit.
Alan:
I've started to see my Exit node restart itself.
are you talking about tor daemon / process restarts or entire host restarts?
How often does tor restart? Does it always restart right after it is relaunched?
I guess you run a RPM based distro. I do not think that the openssl message has anything to do with the restarts. What does your syslog (/var/log/messages) say? any out-of-memory kills?
syslog says it failed but didn't give a reason.
Jun 21 03:07:01 vps110165 systemd: Reloaded Anonymizing overlay network for TCP (multi-instance master). Jun 21 03:07:01 vps110165 systemd: Reloaded Anonymizing overlay network for TCP. Jun 21 03:07:02 vps110165 systemd: tor.service: main process exited, code=exited, status=1/FAILURE Jun 21 03:07:02 vps110165 systemd: Unit tor.service entered failed state. Jun 21 03:07:02 vps110165 systemd: tor.service failed. Jun 21 03:07:03 vps110165 systemd: tor.service holdoff time over, scheduling restart. Jun 21 03:07:03 vps110165 systemd: Starting Anonymizing overlay network for TCP…
Is this also what you see when you run: systemctl status tor.service
Jun 21 03:07:03.000 [warn] Fixing permissions on directory /var/lib/tor Jun 21 03:07:03.000 [notice] We were built to run on a 64-bit CPU, with OpenSSL 1.0.1 or later, but with a version of OpenSSL that apparently lacks accelerated support for the NIST P-224 and P-256 groups. Building openssl with such support (using the enable-ec_nistp_64_gcc_128 option when configuring it) would make ECD H much faster.
Are these the last entries in the tor log?
What do you see when you run: journalctl -xn just after tor has exited?
T
-- Tim Wilson-Brown (teor)
teor2345 at gmail dot com PGP C855 6CED 5D90 A0C5 29F6 4D43 450C BA7F 968F 094B ricochet:ekmygaiu4rzgsk6n xmpp: teor at torproject dot org ------------------------------------------------------------------------
On 22 Jun 2017, at 04:38, Alan tor-relay@clutterbuck.uk wrote:
thanks for running an exit.
Alan:
I've started to see my Exit node restart itself.
are you talking about tor daemon / process restarts or entire host restarts?
How often does tor restart? Does it always restart right after it is relaunched?
3 days running it has restarted and it always restarts. It has done it again today. The servers uptime is 179 days, i will reboot it and see if helps.
I guess you run a RPM based distro. I do not think that the openssl message has anything to do with the restarts. What does your syslog (/var/log/messages) say? any out-of-memory kills?
syslog says it failed but didn't give a reason.
Jun 21 03:07:01 vps110165 systemd: Reloaded Anonymizing overlay network for TCP (multi-instance master). Jun 21 03:07:01 vps110165 systemd: Reloaded Anonymizing overlay network for TCP. Jun 21 03:07:02 vps110165 systemd: tor.service: main process exited, code=exited, status=1/FAILURE Jun 21 03:07:02 vps110165 systemd: Unit tor.service entered failed state. Jun 21 03:07:02 vps110165 systemd: tor.service failed. Jun 21 03:07:03 vps110165 systemd: tor.service holdoff time over, scheduling restart. Jun 21 03:07:03 vps110165 systemd: Starting Anonymizing overlay network for TCPâ¦
Is this also what you see when you run: systemctl status tor.service
[root@vps110165 ~]# systemctl status tor.service ● tor.service - Anonymizing overlay network for TCP Loaded: loaded (/usr/lib/systemd/system/tor.service; disabled; vendor preset: disabled) Active: active (running) since Thu 2017-06-22 03:45:03 EDT; 1h 26min ago Process: 4183 ExecReload=/bin/kill -HUP ${MAINPID} (code=exited, status=0/SUCCESS) Process: 4204 ExecStartPre=/usr/bin/tor --runasdaemon 0 --defaults-torrc /usr/share/tor/defaults-torrc -f /etc/tor/torrc --verify-config (code=exited, status=0/SUCCESS) Main PID: 4205 (tor) CGroup: /system.slice/tor.service └─4205 /usr/bin/tor --runasdaemon 0 --defaults-torrc /usr/share/tor/defaults-torrc -f /etc/tor/torrc
Jun 22 03:45:03 vps110165.vps.ovh.ca tor[4205]: Jun 22 03:45:03.425 [notice] Tor 0.2.9.10 (git-1f6c8eda0073f464) running on Linux with Libevent 2.0....ib 1.2.7. Jun 22 03:45:03 vps110165.vps.ovh.ca tor[4205]: Jun 22 03:45:03.425 [notice] Tor can't help you if you use it wrong! Learn how to be safe at https:/...d#warning Jun 22 03:45:03 vps110165.vps.ovh.ca tor[4205]: Jun 22 03:45:03.425 [notice] Read configuration file "/usr/share/tor/defaults-torrc". Jun 22 03:45:03 vps110165.vps.ovh.ca tor[4205]: Jun 22 03:45:03.425 [notice] Read configuration file "/etc/tor/torrc". Jun 22 03:45:03 vps110165.vps.ovh.ca tor[4205]: Jun 22 03:45:03.428 [notice] Based on detected system memory, MaxMemInQueues is set to 1464 MB. You ... by hand. Jun 22 03:45:03 vps110165.vps.ovh.ca tor[4205]: Jun 22 03:45:03.428 [warn] Tor is running as an exit relay. If you did not want this behavior, pleas...tibility. Jun 22 03:45:03 vps110165.vps.ovh.ca tor[4205]: Jun 22 03:45:03.432 [notice] Opening Socks listener on 127.0.0.1:9050 Jun 22 03:45:03 vps110165.vps.ovh.ca tor[4205]: Jun 22 03:45:03.432 [notice] Opening Control listener on 127.0.0.1:9051 Jun 22 03:45:03 vps110165.vps.ovh.ca tor[4205]: Jun 22 03:45:03.432 [notice] Opening OR listener on 0.0.0.0:8081 Jun 22 03:45:03 vps110165.vps.ovh.ca tor[4205]: Jun 22 03:45:03.433 [notice] Opening Directory listener on 0.0.0.0:8080 Hint: Some lines were ellipsized, use -l to show in full.
Jun 21 03:07:03.000 [warn] Fixing permissions on directory /var/lib/tor Jun 21 03:07:03.000 [notice] We were built to run on a 64-bit CPU, with OpenSSL 1.0.1 or later, but with a version of OpenSSL that apparently lacks accelerated support for the NIST P-224 and P-256 groups. Building openssl with such support (using the enable-ec_nistp_64_gcc_128 option when configuring it) would make ECD H much faster.
Are these the last entries in the tor log?
What do you see when you run: journalctl -xn just after tor has exited?
-- Logs begin at Sat 2016-12-24 06:44:45 EST, end at Thu 2017-06-22 05:01:01 EDT. -- Jun 22 04:01:01 vps110165.vps.ovh.ca systemd[1]: Started Session 4384 of user root. -- Subject: Unit session-4384.scope has finished start-up -- Defined-By: systemd -- Support: http://lists.freedesktop.org/mailman/listinfo/systemd-devel -- -- Unit session-4384.scope has finished starting up. -- -- The start-up result is done. Jun 22 04:01:01 vps110165.vps.ovh.ca systemd[1]: Starting Session 4384 of user root. -- Subject: Unit session-4384.scope has begun start-up -- Defined-By: systemd -- Support: http://lists.freedesktop.org/mailman/listinfo/systemd-devel -- -- Unit session-4384.scope has begun starting up. Jun 22 04:01:01 vps110165.vps.ovh.ca CROND[4210]: (root) CMD (run-parts /etc/cron.hourly) Jun 22 04:01:01 vps110165.vps.ovh.ca run-parts(/etc/cron.hourly)[4213]: starting 0anacron Jun 22 04:01:01 vps110165.vps.ovh.ca run-parts(/etc/cron.hourly)[4219]: finished 0anacron Jun 22 05:01:01 vps110165.vps.ovh.ca systemd[1]: Started Session 4385 of user root. -- Subject: Unit session-4385.scope has finished start-up -- Defined-By: systemd -- Support: http://lists.freedesktop.org/mailman/listinfo/systemd-devel -- -- Unit session-4385.scope has finished starting up. -- -- The start-up result is done. Jun 22 05:01:01 vps110165.vps.ovh.ca systemd[1]: Starting Session 4385 of user root. -- Subject: Unit session-4385.scope has begun start-up -- Defined-By: systemd -- Support: http://lists.freedesktop.org/mailman/listinfo/systemd-devel -- -- Unit session-4385.scope has begun starting up. Jun 22 05:01:01 vps110165.vps.ovh.ca CROND[4224]: (root) CMD (run-parts /etc/cron.hourly) Jun 22 05:01:01 vps110165.vps.ovh.ca run-parts(/etc/cron.hourly)[4227]: starting 0anacron Jun 22 05:01:01 vps110165.vps.ovh.ca run-parts(/etc/cron.hourly)[4233]: finished 0anacron
T
-- Tim Wilson-Brown (teor)
teor2345 at gmail dot com PGP C855 6CED 5D90 A0C5 29F6 4D43 450C BA7F 968F 094B ricochet:ekmygaiu4rzgsk6n xmpp: teor at torproject dot org
tor-relays mailing list tor-relays@lists.torproject.org https://lists.torproject.org/cgi-bin/mailman/listinfo/tor-relays
Hi,
These logs are from a long time after the relay restarted. We need to see the logs at the time the relay stops.
Please look at /var/log/tor/log and send us the 20 lines before the last message that starts like this:
Tor 0.2.9.10 (git-1f6c8eda0073f464) running on Linux with Libevent
It might also help to upgrade to 0.2.9.11, there are a few crash fixes in it.
...
How often does tor restart? Does it always restart right after it is relaunched?
3 days running it has restarted and it always restarts. It has done it again today. The servers uptime is 179 days, i will reboot it and see if helps.
...
T
-- Tim Wilson-Brown (teor)
teor2345 at gmail dot com PGP C855 6CED 5D90 A0C5 29F6 4D43 450C BA7F 968F 094B ricochet:ekmygaiu4rzgsk6n xmpp: teor at torproject dot org ------------------------------------------------------------------------
Hi,
These logs are from a long time after the relay restarted. We need to see the logs at the time the relay stops.
Please look at /var/log/tor/log and send us the 20 lines before the last message that starts like this:
Tor 0.2.9.10 (git-1f6c8eda0073f464) running on Linux with Libevent
It might also help to upgrade to 0.2.9.11, there are a few crash fixes in it.
The warn message is repeated several times before the crash.
Jun 22 05:13:14.000 [warn] Before Tor can create a control socket in "/var/run/tor/control", the directory "/var/run/tor" needs to exist, and to be accessible only by the user account that is running Tor. (On some Unix systems, anybody who can list a socket can connect to it, so Tor is being careful.) Jun 22 05:13:41.000 [notice] Interrupt: we have stopped accepting new connections, and will shut down in 30 seconds. Interrupt again to exit now. Jun 22 05:14:11.000 [notice] Clean shutdown finished. Exiting. Jun 22 05:15:17.000 [notice] Tor 0.2.9.10 (git-1f6c8eda0073f464) opening log file.
...
How often does tor restart? Does it always restart right after it is relaunched?
3 days running it has restarted and it always restarts. It has done it again today. The servers uptime is 179 days, i will reboot it and see if helps.
...
T
-- Tim Wilson-Brown (teor)
teor2345 at gmail dot com PGP C855 6CED 5D90 A0C5 29F6 4D43 450C BA7F 968F 094B ricochet:ekmygaiu4rzgsk6n xmpp: teor at torproject dot org
tor-relays mailing list tor-relays@lists.torproject.org https://lists.torproject.org/cgi-bin/mailman/listinfo/tor-relays
On 22 Jun 2017, at 23:08, Alan tor-relay@clutterbuck.uk wrote:
Hi,
These logs are from a long time after the relay restarted. We need to see the logs at the time the relay stops.
Please look at /var/log/tor/log and send us the 20 lines before the last message that starts like this:
Tor 0.2.9.10 (git-1f6c8eda0073f464) running on Linux with Libevent
It might also help to upgrade to 0.2.9.11, there are a few crash fixes in it.
The warn message is repeated several times before the crash.
Jun 22 05:13:14.000 [warn] Before Tor can create a control socket in "/var/run/tor/control", the directory "/var/run/tor" needs to exist, and to be accessible only by the user account that is running Tor. (On some Unix systems, anybody who can list a socket can connect to it, so Tor is being careful.) Jun 22 05:13:41.000 [notice] Interrupt: we have stopped accepting new connections, and will shut down in 30 seconds. Interrupt again to exit now. Jun 22 05:14:11.000 [notice] Clean shutdown finished. Exiting. Jun 22 05:15:17.000 [notice] Tor 0.2.9.10 (git-1f6c8eda0073f464) opening log file.
What are the permissions on /var/run/tor?
You need to find out which process is sending the interrupt, and why. It could be systemd, because tor hasn't finished launching in a reasonable time.
If it is, you need to find out why tor isn't launching properly. Please send all of the log messages from the time tor starts to the time it stops (it takes a minute or two to stop, right?)
You might have run into a bug like this one with CapabilityBoundingSet: https://trac.torproject.org/projects/tor/ticket/22331 But it's hard to tell without log messages.
T
-- Tim Wilson-Brown (teor)
teor2345 at gmail dot com PGP C855 6CED 5D90 A0C5 29F6 4D43 450C BA7F 968F 094B ricochet:ekmygaiu4rzgsk6n xmpp: teor at torproject dot org ------------------------------------------------------------------------
tor-relays@lists.torproject.org
-
Alan -
nusenu -
Ralph Seichter -
teor